参考文献:
conv_ai
次のコマンドを使用して、このデータセットを TFDS にロードします。
ds = tfds.load('huggingface:conv_ai/conv_ai')
- 説明：
ConvAI is a dataset of human-to-bot conversations labelled for quality. This data can be used to train a metric for evaluating dialogue systems. Moreover, it can be used in the development of chatbots themselves: it contains the information on the quality of utterances and entire dialogues, that can guide a dialogue system in search of better answers.
- ライセンス: 既知のライセンスはありません
- バージョン: 1.0.0
- 分割:
| スプリット | 例 |
|---|---|
'train' | 2778 |
- 特徴：
{
"id": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"dialogId": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"context": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"users": [
{
"userType": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"id": {
"dtype": "string",
"id": null,
"_type": "Value"
}
}
],
"evaluation": [
{
"breadth": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"userId": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"quality": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"engagement": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
],
"thread": [
{
"evaluation": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"text": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"userId": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"time": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
]
}